Weak supervision

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Weak_supervision

Weak supervision (also known as semi-supervised learning) is a paradigm in machine learning, the relevance and notability of which increased with the advent of large language models due to large amount of data required to train them. It is characterized by using a combination of a small amount of human-labeled data (exclusively used in more expensive and time-consuming supervised learning paradigm), followed by a large amount of unlabeled data (used exclusively in unsupervised learning paradigm). In other words, the desired output values are provided only for a subset of the training data. The remaining data is unlabeled or imprecisely labeled. Intuitively, it can be seen as an exam and labeled data as sample problems that the teacher solves for the class as an aid in solving another set of problems. In the transductive setting, these unsolved problems act as exam questions. In the inductive setting, they become practice problems of the sort that will make up the exam. Technically, it could be viewed as performing clustering and then labeling the clusters with the labeled data, pushing the decision boundary away from high-density regions, or learning an underlying one-dimensional manifold where the data reside.

Problem

Tendency for a task to employ supervised vs. unsupervised methods. Task names straddling circle boundaries is intentional. It shows that the classical division of imaginative tasks (left) employing unsupervised methods is blurred in today's learning schemes.

The acquisition of labeled data for a learning problem often requires a skilled human agent (e.g. to transcribe an audio segment) or a physical experiment (e.g. determining the 3D structure of a protein or determining whether there is oil at a particular location). The cost associated with the labeling process thus may render large, fully labeled training sets infeasible, whereas acquisition of unlabeled data is relatively inexpensive. In such situations, semi-supervised learning can be of great practical value. Semi-supervised learning is also of theoretical interest in machine learning and as a model for human learning.

Technique

See also: Active learning (machine learning)

An example of the influence of unlabeled data in semi-supervised learning. The top panel shows a decision boundary we might adopt after seeing only one positive (white circle) and one negative (black circle) example. The bottom panel shows a decision boundary we might adopt if, in addition to the two labeled examples, we were given a collection of unlabeled data (gray circles).

More formally, semi-supervised learning assumes a set of ${\displaystyle l}$ independently identically distributed examples ${\displaystyle x_1,\dots ,x_l\in X}$ with corresponding labels ${\displaystyle y_1,\dots ,y_l\in Y}$ and ${\displaystyle u}$ unlabeled examples ${\displaystyle x_{l+1},\dots ,x_{l+u}\in X}$ are processed. Semi-supervised learning combines this information to surpass the classification performance that can be obtained either by discarding the unlabeled data and doing supervised learning or by discarding the labels and doing unsupervised learning.

Semi-supervised learning may refer to either transductive learning or inductive learning. The goal of transductive learning is to infer the correct labels for the given unlabeled data ${\displaystyle x_{l+1},\dots ,x_{l+u}}$ only. The goal of inductive learning is to infer the correct mapping from ${\displaystyle X}$ to ${\displaystyle Y}$.

It is unnecessary (and, according to Vapnik's principle, imprudent) to perform transductive learning by way of inferring a classification rule over the entire input space; however, in practice, algorithms formally designed for transduction or induction are often used interchangeably.

Assumptions

In order to make any use of unlabeled data, some relationship to the underlying distribution of data must exist. Semi-supervised learning algorithms make use of at least one of the following assumptions:

Continuity / smoothness assumption

Points that are close to each other are more likely to share a label. This is also generally assumed in supervised learning and yields a preference for geometrically simple decision boundaries. In the case of semi-supervised learning, the smoothness assumption additionally yields a preference for decision boundaries in low-density regions, so few points are close to each other but in different classes.

Cluster assumption

The data tend to form discrete clusters, and points in the same cluster are more likely to share a label (although data that shares a label may spread across multiple clusters). This is a special case of the smoothness assumption and gives rise to feature learning with clustering algorithms.

Manifold assumption

Main article: Manifold hypothesis

The data lie approximately on a manifold of much lower dimension than the input space. In this case learning the manifold using both the labeled and unlabeled data can avoid the curse of dimensionality. Then learning can proceed using distances and densities defined on the manifold.

The manifold assumption is practical when high-dimensional data are generated by some process that may be hard to model directly, but which has only a few degrees of freedom. For instance, human voice is controlled by a few vocal folds, and images of various facial expressions are controlled by a few muscles. In these cases, it is better to consider distances and smoothness in the natural space of the generating problem, rather than in the space of all possible acoustic waves or images, respectively.

History

The heuristic approach of self-training (also known as self-learning or self-labeling) is historically the oldest approach to semi-supervised learning, with examples of applications starting in the 1960s.

The transductive learning framework was formally introduced by Vladimir Vapnik in the 1970s. Interest in inductive learning using generative models also began in the 1970s. A probably approximately correct learning bound for semi-supervised learning of a Gaussian mixture was demonstrated by Ratsaby and Venkatesh in 1995.

Methods

Generative models

Generative approaches to statistical learning first seek to estimate ${\displaystyle p(x|y)}$, the distribution of data points belonging to each class. The probability ${\displaystyle p(y|x)}$ that a given point ${\displaystyle x}$ has label ${\displaystyle y}$ is then proportional to ${\displaystyle p(x|y)p(y)}$ by Bayes' rule. Semi-supervised learning with generative models can be viewed either as an extension of supervised learning (classification plus information about ${\displaystyle p(x)}$) or as an extension of unsupervised learning (clustering plus some labels).

Generative models assume that the distributions take some particular form ${\displaystyle p(x|y,\theta )}$ parameterized by the vector ${\displaystyle \theta }$. If these assumptions are incorrect, the unlabeled data may actually decrease the accuracy of the solution relative to what would have been obtained from labeled data alone. However, if the assumptions are correct, then the unlabeled data necessarily improves performance.

The unlabeled data are distributed according to a mixture of individual-class distributions. In order to learn the mixture distribution from the unlabeled data, it must be identifiable, that is, different parameters must yield different summed distributions. Gaussian mixture distributions are identifiable and commonly used for generative models.

The parameterized joint distribution can be written as ${\displaystyle p(x,y|\theta )=p(y|\theta )p(x|y,\theta )}$ by using the chain rule. Each parameter vector ${\displaystyle \theta }$ is associated with a decision function ${\displaystyle f_{\theta }(x)=\underset{y}{\mathrm{argmax}}p(y|x,\theta )}$. The parameter is then chosen based on fit to both the labeled and unlabeled data, weighted by ${\displaystyle \lambda }$:

${\displaystyle \underset{\mathrm{\Theta }}{\mathrm{argmax}}\left(\log p(\{x_i,y_i{\}}_{i=1}^l|\theta )+\lambda \log p(\{x_i{\}}_{i=l+1}^{l+u}|\theta )\right)}$

Low-density separation

Another major class of methods attempts to place boundaries in regions with few data points (labeled or unlabeled). One of the most commonly used algorithms is the transductive support vector machine, or TSVM (which, despite its name, may be used for inductive learning as well). Whereas support vector machines for supervised learning seek a decision boundary with maximal margin over the labeled data, the goal of TSVM is a labeling of the unlabeled data such that the decision boundary has maximal margin over all of the data. In addition to the standard hinge loss ${\displaystyle (1-yf(x){)}_{+}}$ for labeled data, a loss function ${\displaystyle (1-|f(x)|{)}_{+}}$ is introduced over the unlabeled data by letting ${\displaystyle y=\mathrm{sign}f(x)}$. TSVM then selects ${\displaystyle f^{\ast }(x)=h^{\ast }(x)+b}$ from a reproducing kernel Hilbert space ${\displaystyle \mathcal{H}}$ by minimizing the regularized empirical risk:

${\displaystyle f^{\ast }=\underset{f}{\mathrm{argmin}}\left({\displaystyle \sum _{i=1}^l(1-y_{i}f(x_i){)}_{+}+{\lambda }_1\Vert h{\Vert }_{\mathcal{H}}^2+{\lambda }_2\sum _{i=l+1}^{l+u}(1-|f(x_i)|{)}_{+}}\right)}$

An exact solution is intractable due to the non-convex term ${\displaystyle (1-|f(x)|{)}_{+}}$, so research focuses on useful approximations.

Other approaches that implement low-density separation include Gaussian process models, information regularization, and entropy minimization (of which TSVM is a special case).

Laplacian regularization

Laplacian regularization has been historically approached through graph-Laplacian. Graph-based methods for semi-supervised learning use a graph representation of the data, with a node for each labeled and unlabeled example. The graph may be constructed using domain knowledge or similarity of examples; two common methods are to connect each data point to its ${\displaystyle k}$ nearest neighbors or to examples within some distance ${\displaystyle ϵ}$. The weight ${\displaystyle W_{ij}}$ of an edge between ${\displaystyle x_i}$ and ${\displaystyle x_j}$ is then set to ${\displaystyle e^{-\Vert x_i-x_j{\Vert }^2/{ϵ}^2}}$.

Within the framework of manifold regularization, the graph serves as a proxy for the manifold. A term is added to the standard Tikhonov regularization problem to enforce smoothness of the solution relative to the manifold (in the intrinsic space of the problem) as well as relative to the ambient input space. The minimization problem becomes

${\displaystyle \underset{f\in \mathcal{H}}{\mathrm{argmin}}\left(\frac{1}{l}{\displaystyle \sum _{i=1}^{l}V(f(x_i),y_i)+{\lambda }_A\Vert f{\Vert }_{\mathcal{H}}^2+{\lambda }_I{\int }_{\mathcal{M}}\Vert {\mathrm{\nabla }}_{\mathcal{M}}f(x){\Vert }^{2}dp(x)}\right)}$

where ${\displaystyle \mathcal{H}}$ is a reproducing kernel Hilbert space and ${\displaystyle \mathcal{M}}$ is the manifold on which the data lie. The regularization parameters ${\displaystyle {\lambda }_A}$ and ${\displaystyle {\lambda }_I}$ control smoothness in the ambient and intrinsic spaces respectively. The graph is used to approximate the intrinsic regularization term. Defining the graph Laplacian ${\displaystyle L=D-W}$ where ${\displaystyle D_{ii}=\sum _{j=1}^{l+u}{W}_{ij}}$ and ${\displaystyle \mathbf{f}}$ is the vector ${\displaystyle [f(x_1)\dots f(x_{l+u})]}$, we have

${\displaystyle {\mathbf{f}}^{T}L\mathbf{f}={\displaystyle \sum _{i,j=1}^{l+u}{W}_{ij}(f_i-f_j{)}^2\approx {\int }_{\mathcal{M}}\Vert {\mathrm{\nabla }}_{\mathcal{M}}f(x){\Vert }^{2}dp(x)}}$.

The graph-based approach to Laplacian regularization is to put in relation with finite difference method.

The Laplacian can also be used to extend the supervised learning algorithms: regularized least squares and support vector machines (SVM) to semi-supervised versions Laplacian regularized least squares and Laplacian SVM.

Heuristic approaches

Some methods for semi-supervised learning are not intrinsically geared to learning from both unlabeled and labeled data, but instead make use of unlabeled data within a supervised learning framework. For instance, the labeled and unlabeled examples ${\displaystyle x_1,\dots ,x_{l+u}}$ may inform a choice of representation, distance metric, or kernel for the data in an unsupervised first step. Then supervised learning proceeds from only the labeled examples. In this vein, some methods learn a low-dimensional representation using the supervised data and then apply either low-density separation or graph-based methods to the learned representation. Iteratively refining the representation and then performing semi-supervised learning on said representation may further improve performance.

Self-training is a wrapper method for semi-supervised learning. First a supervised learning algorithm is trained based on the labeled data only. This classifier is then applied to the unlabeled data to generate more labeled examples as input for the supervised learning algorithm. Generally only the labels the classifier is most confident in are added at each step. In natural language processing, a common self-training algorithm is the Yarowsky algorithm for problems like word sense disambiguation, accent restoration, and spelling correction.

Co-training is an extension of self-training in which multiple classifiers are trained on different (ideally disjoint) sets of features and generate labeled examples for one another.

In human cognition

Human responses to formal semi-supervised learning problems have yielded varying conclusions about the degree of influence of the unlabeled data. More natural learning problems may also be viewed as instances of semi-supervised learning. Much of human concept learning involves a small amount of direct instruction (e.g. parental labeling of objects during childhood) combined with large amounts of unlabeled experience (e.g. observation of objects without naming or counting them, or at least without feedback).

Human infants are sensitive to the structure of unlabeled natural categories such as images of dogs and cats or male and female faces. Infants and children take into account not only unlabeled examples, but the sampling process from which labeled examples arise.

Evolutionary algorithm

From Wikipedia, the free encyclopedia
https://en.wikipedia.org/wiki/Evolutionary_algorithm

In computational intelligence (CI), an evolutionary algorithm (EA) is a subset of evolutionary computation, a generic population-based metaheuristic optimization algorithm. An EA uses mechanisms inspired by biological evolution, such as reproduction, mutation, recombination, and selection. Candidate solutions to the optimization problem play the role of individuals in a population, and the fitness function determines the quality of the solutions (see also loss function). Evolution of the population then takes place after the repeated application of the above operators.

Evolutionary algorithms often perform well approximating solutions to all types of problems because they ideally do not make any assumption about the underlying fitness landscape. Techniques from evolutionary algorithms applied to the modeling of biological evolution are generally limited to explorations of microevolutionary processes and planning models based upon cellular processes. In most real applications of EAs, computational complexity is a prohibiting factor. In fact, this computational complexity is due to fitness function evaluation. Fitness approximation is one of the solutions to overcome this difficulty. However, seemingly simple EA can solve often complex problems; therefore, there may be no direct link between algorithm complexity and problem complexity.

Implementation

The following is an example of a generic single-objective genetic algorithm.

Step One: Generate the initial population of individuals randomly. (First generation)

Step Two: Repeat the following regenerational steps until termination (time limit, sufficient fitness achieved, etc.):

1. Evaluate the fitness of each individual in the population
2. Select the individuals for reproduction based on their fitness. (Parents)
3. Breed new individuals through crossover and mutation operations to give birth to offspring.
4. Replace the least-fit individuals of the population with new individuals.

Types

Similar techniques differ in genetic representation and other implementation details, and the nature of the particular applied problem.

• Genetic algorithm – This is the most popular type of EA. One seeks the solution of a problem in the form of strings of numbers (traditionally binary, although the best representations are usually those that reflect something about the problem being solved), by applying operators such as recombination and mutation (sometimes one, sometimes both). This type of EA is often used in optimization problems.
• Genetic programming – Here the solutions are in the form of computer programs, and their fitness is determined by their ability to solve a computational problem. There are many variants of Genetic Programming, including Cartesian genetic programming, gene expression programming, grammatical evolution, linear genetic programming, multi expression programming etc.
• Evolutionary programming – Similar to genetic programming, but the structure of the program is fixed and its numerical parameters are allowed to evolve.
• Evolution strategy – Works with vectors of real numbers as representations of solutions, and typically uses self-adaptive mutation rates. The method is mainly used for numerical optimization, although there are also variants for combinatorial tasks.
• Differential evolution – Based on vector differences and is therefore primarily suited for numerical optimization problems.
• Coevolutionary algorithm – Similar to genetic algorithms and evolution strategies, but the created solutions are compared on the basis of their outcomes from interactions with other solutions. Solutions can either compete or cooperate during the search process. Coevolutionary algorithms are often used in scenarios where the fitness landscape is dynamic, complex, or involves competitive interactions.
• Neuroevolution – Similar to genetic programming but the genomes represent artificial neural networks by describing structure and connection weights. The genome encoding can be direct or indirect.
• Learning classifier system – Here the solution is a set of classifiers (rules or conditions). A Michigan-LCS evolves at the level of individual classifiers whereas a Pittsburgh-LCS uses populations of classifier-sets. Initially, classifiers were only binary, but now include real, neural net, or S-expression types. Fitness is typically determined with either a strength or accuracy based reinforcement learning or supervised learning approach.
• Quality–Diversity algorithms – QD algorithms simultaneously aim for high-quality and diverse solutions. Unlike traditional optimization algorithms that solely focus on finding the best solution to a problem, QD algorithms explore a wide variety of solutions across a problem space and keep those that are not just high performing, but also diverse and unique.

Theoretical background

The following theoretical principles apply to all or almost all EAs.

No free lunch theorem

The no free lunch theorem of optimization states that all optimization strategies are equally effective when the set of all optimization problems is considered. Under the same condition, no evolutionary algorithm is fundamentally better than another. This can only be the case if the set of all problems is restricted. This is exactly what is inevitably done in practice. Therefore, to improve an EA, it must exploit problem knowledge in some form (e.g. by choosing a certain mutation strength or a problem-adapted coding). Thus, if two EAs are compared, this constraint is implied. In addition, an EA can use problem specific knowledge by, for example, not randomly generating the entire start population, but creating some individuals through heuristics or other procedures. Another possibility to tailor an EA to a given problem domain is to involve suitable heuristics, local search procedures or other problem-related procedures in the process of generating the offspring. This form of extension of an EA is also known as a memetic algorithm. Both extensions play a major role in practical applications, as they can speed up the search process and make it more robust.

Convergence

Further information: Convergence (evolutionary computing)

See also: Convergent evolution

For EAs in which, in addition to the offspring, at least the best individual of the parent generation is used to form the subsequent generation (so-called elitist EAs), there is a general proof of convergence under the condition that an optimum exists. Without loss of generality, a maximum search is assumed for the proof:

From the property of elitist offspring acceptance and the existence of the optimum it follows that per generation ${\displaystyle k}$ an improvement of the fitness ${\displaystyle F}$ of the respective best individual ${\displaystyle x^{\prime }}$ will occur with a probability ${\displaystyle P>0}$. Thus:

${\displaystyle F(x_1^{\prime })\le F(x_2^{\prime })\le F(x_3^{\prime })\le \cdots \le F(x_k^{\prime })\le \cdots }$

I.e., the fitness values represent a monotonically non-decreasing sequence, which is bounded due to the existence of the optimum. From this follows the convergence of the sequence against the optimum.

Since the proof makes no statement about the speed of convergence, it is of little help in practical applications of EAs. But it does justify the recommendation to use elitist EAs. However, when using the usual panmictic population model, elitist EAs tend to converge prematurely more than non-elitist ones. In a panmictic population model, mate selection (step 2 of the section about implementation) is such that every individual in the entire population is eligible as a mate. In non-panmictic populations, selection is suitably restricted, so that the dispersal speed of better individuals is reduced compared to panmictic ones. Thus, the general risk of premature convergence of elitist EAs can be significantly reduced by suitable population models that restrict mate selection.

Virtual alphabets

With the theory of virtual alphabets, David E. Goldberg showed in 1990 that by using a representation with real numbers, an EA that uses classical recombination operators (e.g. uniform or n-point crossover) cannot reach certain areas of the search space, in contrast to a coding with binary numbers. This results in the recommendation for EAs with real representation to use arithmetic operators for recombination (e.g. arithmetic mean or intermediate recombination). With suitable operators, real-valued representations are more effective than binary ones, contrary to earlier opinion.

Comparison to biological processes

A possible limitation of many evolutionary algorithms is their lack of a clear genotype–phenotype distinction. In nature, the fertilized egg cell undergoes a complex process known as embryogenesis to become a mature phenotype. This indirect encoding is believed to make the genetic search more robust (i.e. reduce the probability of fatal mutations), and also may improve the evolvability of the organism.  Such indirect (also known as generative or developmental) encodings also enable evolution to exploit the regularity in the environment. Recent work in the field of artificial embryogeny, or artificial developmental systems, seeks to address these concerns. And gene expression programming successfully explores a genotype–phenotype system, where the genotype consists of linear multigenic chromosomes of fixed length and the phenotype consists of multiple expression trees or computer programs of different sizes and shapes.

Comparison to Monte-Carlo methods

Both method classes have in common that their individual search steps are determined by chance. The main difference, however, is that EAs, like many other metaheuristics, learn from past search steps and incorporate this experience into the execution of the next search steps in a method-specific form. With EAs, this is done firstly through the fitness-based selection operators for partner choice and the formation of the next generation. And secondly, in the type of search steps: In EA, they start from a current solution and change it or they mix the information of two solutions. In contrast, when dicing out new solutions in Monte-Carlo methods, there is usually no connection to existing solutions.

If, on the other hand, the search space of a task is such that there is nothing to learn, Monte-Carlo methods are an appropriate tool, as they do not contain any algorithmic overhead that attempts to draw suitable conclusions from the previous search. An example of such tasks is the proverbial search for a needle in a haystack, e.g. in the form of a flat (hyper)plane with a single narrow peak.

Applications

The areas in which evolutionary algorithms are practically used are almost unlimited and range from industry, engineering, complex scheduling, agriculture, robot movement planning and finance to research and art. The application of an evolutionary algorithm requires some rethinking from the inexperienced user, as the approach to a task using an EA is different from conventional exact methods and this is usually not part of the curriculum of engineers or other disciplines. For example, the fitness calculation must not only formulate the goal but also support the evolutionary search process towards it, e.g. by rewarding improvements that do not yet lead to a better evaluation of the original quality criteria. For example, if peak utilisation of resources such as personnel deployment or energy consumption is to be avoided in a scheduling task, it is not sufficient to assess the maximum utilisation. Rather, the number and duration of exceedances of a still acceptable level should also be recorded in order to reward reductions below the actual maximum peak value. There are therefore some publications that are aimed at the beginner and want to help avoiding beginner's mistakes as well as leading an application project to success. This includes clarifying the fundamental question of when an EA should be used to solve a problem and when it is better not to.

Related techniques and other global search methods

There are some other proven and widely used methods of nature inspired global search techniques such as

• Memetic algorithm – A hybrid method, inspired by Richard Dawkins's notion of a meme. It commonly takes the form of a population-based algorithm (frequently an EA) coupled with individual learning procedures capable of performing local refinements. Emphasizes the exploitation of problem-specific knowledge and tries to orchestrate local and global search in a synergistic way.
• A celular evolutionary or memetic algorithm uses a topological neighbouhood relation between the individuals of a population for restricting the mate selection and by that reducing the propagation speed of above-average individuals. The idea is to maintain genotypic diversity in the poulation over a longer period of time to reduce the risk of premature convergence.
• Ant colony optimization is based on the ideas of ant foraging by pheromone communication to form paths. Primarily suited for combinatorial optimization and graph problems.
• Particle swarm optimization is based on the ideas of animal flocking behaviour. Also primarily suited for numerical optimization problems.
• Gaussian adaptation – Based on information theory. Used for maximization of manufacturing yield, mean fitness or average information. See for instance Entropy in thermodynamics and information theory.

In addition, many new nature-inspired or methaphor-guided algorithms have been proposed since the beginning of this century. For criticism of most publications on these, see the remarks at the end of the introduction to the article on metaheuristics.

Examples

In 2020, Google stated that their AutoML-Zero can successfully rediscover classic algorithms such as the concept of neural networks.

The computer simulations Tierra and Avida attempt to model macroevolutionary dynamics.

Religious abuse

From Wikipedia, the free encyclopedia
https://en.wikipedia.org/wiki/Religious_abuse

Religious abuse is abuse administered through religion, including harassment, humiliation, spiritual abuse or religious violence. Religious abuse may also include the misuse of religion for selfish, secular, or ideological ends, such as the abuse of a clerical position.

Religious abuse can be perpetuated by religious leaders or other members of a religious community, and it can happen in any religion or faith. Some examples of religious abuse include using religious teachings to justify abuse, enforcing strict religious rules and practices that are harmful, shaming or ostracizing individuals who do not conform to religious norms, using religious authority to manipulate or control others, and denying access to medical care or other basic needs in the name of religion.

Religious abuse can have serious and long-lasting effects on individuals and communities, including psychological trauma, emotional distress, loss of faith, and even physical harm. It is important for individuals and religious communities to be aware of the signs of religious abuse and to take steps to prevent it from happening.

Psychological

One specific meaning of the term religious abuse refers to psychological manipulation and harm inflicted on a person by using the teachings of their religion. This is perpetrated by members of the same or similar faith that includes the use of a position of authority within the religion. It is most often directed at children and emotionally vulnerable adults, and the motivations behind such abuse vary, and can be either well-intentioned or malicious.

Even well-intentioned religious abuse can have long-term psychological consequences, such as the victim developing phobias or long-term depression. They may have a sense of shame that persists even after they leave the religion. A person can also be manipulated into avoiding a beneficial action (such as a medical treatment) or to engage in a harmful behavior.

In his book Religious Abuse, pastor Keith Wright describes an example of such abuse. When he was a child, his Christian Scientist mother became very ill and eventually was convinced to seek medical treatment at an inpatient facility. Members of her church went to the treatment center to convince her to trust prayer rather than treatment, and to leave. She died shortly thereafter. While the church members may not have had any malicious intent, their use of their religion's teachings to manipulate Wright's mother ultimately resulted in her death.

A more recent study among 200 university students has shown that 12.5% of students reported being victimized by at least one form of religious or ritual abuse (RA). The study, which was published in the Journal of Interpersonal Violence, showed that religious/ritual abuse may result in mental health issues such as dissociative disorders.

Religious morality can in some cases contribute to victim blaming.

Against children

See also: Sexual grooming § Religious settings

Religiously-based psychological abuse of children can involve using teachings to subjugate children through fear, or indoctrinating the child in the beliefs of their particular religion whilst suppressing other perspectives. Psychologist Jill Mytton describes this as crushing the child's chance to form a personal morality and belief system; it makes them utterly reliant on their religion or parents, and they never learn to reflect critically on the information they receive. Similarly, the use of fear and a judgmental environment (such as the concept of Hell) to control the child can be traumatic.

Spiritual

Spiritual abuse includes:

• Psychological abuse and emotional abuse
• Physical abuse including physical injury (e.g., tatbir) and deprivation of sustenance.
• Sexual abuse
• Any act by deeds or words that shame or diminish the dignity of a person.
• Intimidation and the requirement to submit to a spiritual authority without any right to dissent.
• Unreasonable control of a person's basic right to exercise freewill in spiritual or natural matters.
• False accusations and repeated criticism by labeling a person as, for example, disobedient, rebellious, lacking faith, demonized, apostate, an enemy of the church or of a deity.
• Isolationism, separation, disenfranchisement or estrangement from family and friends outside the group due to cult-religious or spiritual or indigenous beliefs.
• Esotericism, hidden agendas and requirements revealed to members only as they successfully advance through various stages of a faith.
• Enforced practice of spiritualism, mysticism, or other ideologies peculiar to members of that religion.
• Financial exploitation or enslavement of adherents.

Background

The term spiritual abuse was purportedly coined in the late twentieth century to refer to alleged abuse of authority by church leaders, albeit some scholars and historians would dispute that claim, citing prior literary appearances of the term in literature on religion and psychology. Lambert defines spiritual abuse as "a type of psychological predomination that could be rightly termed—religious enslavement". He further identifies "religious enslavement" as being a product of what is termed in the Bible "witchcraft" or "sorcery". A key element of the experience of spiritual abuse is the perceived 'divine position' of the abuser which leads to an environment of infallibility.

Characteristics

Ronald Enroth in Churches That Abuse identifies five categories:

1. Authority and power: abuse arises when leaders of a group arrogate to themselves power and authority that lacks the dynamics of open accountability and the capacity to question or challenge decisions made by leaders. The shift entails moving from general respect for an office bearer to one where members loyally submit without any right to dissent.
2. Manipulation and control: abusive groups are characterized by social dynamics where fear, guilt or threats are routinely used to produce unquestioning obedience, group conformity or stringent tests of loyalty. The leader-disciple relationship may become one in which the leader's decisions control and usurp the disciple's right or capacity to make choices.
3. Elitism and persecution: abusive groups depict themselves as unique and have a strong organizational tendency to be separate from other bodies and institutions. The social dynamism of the group involves being independent or separate, with diminishing possibilities for internal correction, reflection, or external criticism.
4. Life-style and experience: abusive groups foster rigidity in behavior and belief that requires conformity to the group's ideals.
5. Dissent and discipline: abusive groups tend to suppress any kind of internal challenge to decisions made by leaders.

Agnes and John Lawless argue in The Drift into Deception that there are eight characteristics of spiritual abuse, and some of these clearly overlap with Enroth's criteria. They list the eight marks of spiritual abuse as comprising:

1. Charisma and pride
2. Anger and intimidation
3. Greed and fraud
4. Immorality
5. Enslaving authoritarian structure
6. Exclusivity
7. Demanding loyalty and honor
8. New revelation

The author of Charismatic Captivation, Steven Lambert, in a post on the book's website delineates "33 Signs of Spiritual Abuse", including:

1. Apotheosis or de facto deification of the leadership.
2. Absolute authority of the leadership.
3. Pervasive abuse and misuse of authority in personal dealings with members to coerce submission.
4. Paranoia, inordinate egotism or narcissism, and insecurity by the leaders.
5. Abuse and inordinate incidence of "church discipline" particularly in matters not expressly considered to be church discipline issues.
6. Inordinate attention to maintaining the public image of the ministry and lambasting of all "critics".
7. Constant indoctrination with a "group" or "family" mentality that impels members to exalt the corporate "life" and goals of the church-group over their personal goals, callings, objectives or relationships.
8. Members are psychologically traumatized, terrorized and indoctrinated with numerous fears aimed at creating an over-dependence or codependence on their leaders and the corporate group.
9. Members may be required to obtain the approval (or witness) of their leader(s) for decisions regarding personal matters.
10. Frequent preaching from the pulpit discouraging leaving the religion or disobeying the leaderships' dictates.
11. Members departing without the blessing of the leadership do so under a cloud of suspicion, shame, or slander.
12. Departing members often suffer from psychological problems and display the symptoms associated with post-traumatic stress disorder (PTSD).

Research and examples

Flavil Yeakley's team of researchers conducted field-tests with members of the Boston Church of Christ using the Myers-Briggs Type Indicator. In The Discipling Dilemma Yeakley reports that the members tested "showed a high level of change in psychological type scores", with a "clear pattern of convergence in a single type". The same tests were conducted on five mainline denominations and with six groups that are popularly labeled as cults or manipulative sects. Yeakley's test results showed that the pattern in the Boston Church "was not found among other churches of Christ or among members of five mainline denominations, but that it was found in studies of six manipulative sects." The research did not show that the Boston Church was "attracting people with a psychological need for high levels of control", but Yeakley concluded that "they are producing conformity in psychological type" which he deemed to be "unnatural, unhealthy and dangerous."

This was not a longitudinal study and relied on asking participants to answer the survey three times; once as they imagined they might answer five years prior, once as their present selves and once as they imagined they might answer after five years of influence in the sect. The author insists that despite this, "any significant changes in the pattern of these perceptions would indicate some kind of group pressure. A high degree of change and a convergence in a single type would be convincing proof that the Boston Church of Christ has some kind of group dynamic operating that tends to produce conformity to the group norm." However it could instead indicate a desire on the part of the respondents to change in the direction indicated. To determine actual changes in MBTI results would require a longitudinal study, since the methodology here was inherently suggestive of its conclusion. This is also amply borne out in its instructions: "The instructions stated clearly that no one was telling them that their answers ought to change. The instructions said that the purpose of the study was simply to find out if there were any changes and, if so, what those changes might indicate."

Physical

Physical abuse in a religious context can take the form of beatings, illegal confinement, neglect, near drowning or even murder in the belief that the child is possessed by evil spirits, practicing sorcery or witchcraft, or has committed some kind of sin that warrants punishment. Such extreme cases are, though, rare.

In 2012, the United Kingdom's Department for Children, Schools and Families instituted a new action plan to investigate the issue of faith-based abuse after several high-profile murders, such as that of Kristy Bamu. Over a term of 10 years, Scotland Yard conducted 83 investigations into allegations of abuse with faith-based elements and feared there were even more that were unreported.

Violence

Main article: Religious violence

Religious violence and extremism (also called communal violence) is a term that covers all phenomena where religion is either the subject or object of violent behavior.

Human sacrifice

Main articles: Human sacrifice, Child sacrifice, and Child sacrifice in pre-Columbian cultures

Human sacrifice (sometimes called ritual murder), has been practiced on a number of different occasions and in many different cultures. The various rationales behind human sacrifice are the same that motivate religious sacrifice in general. Human sacrifice is typically intended to bring good fortune and to pacify the gods. Fertility was another common theme in ancient religious sacrifices.

Human sacrifice may be a ritual practiced in a stable society, and may even be conducive to enhance societal unity, both by creating a bond unifying the sacrificing community, and in combining human sacrifice and capital punishment, by removing individuals that have a perceived negative effect on societal stability (criminals, religious heretics, foreign slaves or prisoners of war). However, outside of civil religion, human sacrifice may also result in outbursts of blood frenzy and mass killings that destabilize society.

Archaeology has uncovered physical evidence of child sacrifice at several locations. Some of the best attested examples are the diverse rites which were part of the religious practices in Mesoamerica and the Inca Empire. Psychologists Alice Miller and Robert Godwin, psychohistorian Lloyd deMause and other advocates of children's rights have written about pre-Columbian sacrifice within the framework of child abuse.

Plutarch (c.46–120 AD) mentions the Carthaginian's ritual burning of small children, as do Tertullian, Orosius, Diodorus Siculus and Philo. Livy and Polybius do not. The Hebrew Bible also mentions what appears to be child sacrifice practised at a place called the Tophet (roasting place) by the Canaanites, and by some Israelites.

Children were thrown to the sharks in ancient Hawaii.

Sacrificial victims were often infants. "The slaughtering of newborn babies may be considered a common event in many cultures" including the Eskimo, the Polynesians, the Ancient Egyptians, the Chinese, the Scandinavians, and various indigenous peoples of Africa, the Americas and Australia.

Initiation rites

Main article: Initiation rite

Artificial deformation of the skull predates written history and dates back as far as 45,000 BCE, as evidenced by two Neanderthal skulls found in Shanidar Cave. It was usually started just after birth and continued until the desired shape was achieved. It may have played a key role in Egyptian and Mayan societies.

In China some boys were castrated, with both the penis and scrotum cut. Other ritual actions have been described by anthropologists. Géza Róheim wrote about initiation rituals performed by Australian natives in which adolescent initiates were forced to drink blood.

Modern practices

In the rituals of some tribes in Papua New Guinea, an elder "picks out a sharp stick of cane and sticks it deep inside a boy's nostrils until he bleeds profusely into the stream of a pool, an act greeted by loud war cries." Afterwards, when boys are initiated into puberty and manhood, they are expected to perform fellatio on the elders. "Not all initiates will participate in this ceremonial homosexual activity but, about five days later, several will have to perform fellatio several times."

Individual cases of ritual murder have been recorded in Brazil, the United States, and Singapore (Toa Payoh ritual murders).

Witch-hunts

Further information: Witch-hunt

Further information: Witchcraft accusations against children in Africa

To this day, witch hunts, trials and accusations are still a real danger in some parts of the world. Trials result in violence against men, women and children, including murder. In The Gambia, about 1,000 people accused of being witches were locked in government detention centers in March 2009, being beaten, forced to drink an unknown hallucinogenic potion, and confess to witchcraft, according to Amnesty International. In Tanzania thousands of elderly Tanzanian women have been strangled, knifed to death and burned alive over the last two decades after being denounced as witches. Ritualistic abuse may also involve children accused of, and punished for, being purported witches in some Central African areas. A child may be blamed for the illness of a relative, for example. Other examples include Ghana, where alleged witches were banished to refugee camps, and the beating and isolation of children accused of being witches in Angola.

Psychohistorical explanation

Main article: Psychohistorical views on infanticide

A small number of academics subscribe to the theory of psychohistory and attribute the abusive rituals to the psychopathological projection of the perpetrators, especially the parents.

This psychohistorical model claims that practices of tribal societies sometimes included incest and the sacrifice, mutilation, rape and torture of children, and that such activities were culturally acceptable.

Survivors

Survivors of religious abuse can develop symptoms of post-traumatic stress disorder (PTSD) in response to their abusive religious experiences. Dr. Marlene Winell, a psychologist and former fundamentalist, coined the term religious trauma syndrome (RTS) in a 2011 article she wrote for the British Association for Behavioural and Cognitive Psychotherapies. Winell describes RTS as "the condition experienced by people who are struggling with leaving an authoritarian, dogmatic religion and coping with the damage of indoctrination."

In the article, Winell identifies four categories of dysfunction: cognitive, affective, functional, and social/cultural.

• Cognitive: Confusion, difficulty with decision-making and critical thinking, dissociation, identity confusion
• Affective: Anxiety, panic attacks, depression, suicidal ideation, anger, grief, guilt, loneliness, lack of meaning
• Functional: Sleep and eating disorders, nightmares, sexual dysfunction, substance abuse, somatization
• Social/cultural: Rupture of family and social network, employment issues, financial stress, problems acculturating into society, interpersonal dysfunction

These symptoms can occur for people who have simply participated in dogmatic expressions of religion, such as fundamentalism. Extreme cases of religious abuse such as authoritarian cult membership, clergy sexual abuse, or mind control tactics used to extremes like the mass suicide at Jonestown may attract public scrutiny. However, individuals can experience chronic religious abuse in the subtle mind-control mechanics of fundamentalism that leads to trauma. While many extreme traumatic experiences associated with religion can cause simple PTSD, scholars are now arguing that chronic abuse through mind control tactics used in fundamentalist settings, whether intentional or not, can induce C-PTSD or developmental trauma.

Exposure therapy or staying in religiously abusive settings may not be conducive to healing for survivors of religious abuse. Healing can come through support groups, therapy, and psychoeducation. In some cultures, survivors have opportunities to recover and live vibrant lives after they leave religiously abusive settings.

Psychohistory

 

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Psychohistory

Psychohistory is an amalgam of psychology, history, and related social sciences and the humanities. Its proponents claim to examine the "why" of history, especially the difference between stated intention and actual behavior. It works to combine the insights of psychology, especially psychoanalysis, with the research methodology of the social sciences and humanities to understand the emotional origin of the behavior of individuals, groups and nations, past and present. Work in the field has been done in the areas of childhood, creativity, dreams, family dynamics, overcoming adversity, personality, political and presidential psychobiography. There are major psychohistorical studies of anthropology, art, ethnology, history, politics and political science, and much else.

Description

Rembrandt's painting of the sacrifice of Isaac (Gen.22). Psychohistory holds that ritual child sacrifice once occurred in most cultures.

Psychohistorians claim to derive many of its concepts from areas that are perceived to be ignored by conventional historians and anthropologists as shaping factors of human history, in particular, the effects of parenting practice and child abuse. According to conventional historians "the science of culture is independent of the laws of biology and psychology" and "the determining cause of a social fact should be sought among social facts preceding and not among the states of individual consciousness".

Psychohistorians, on the other hand, suggest that social behavior such as crime and war may be a self-destructive re-enactment of earlier abuse and neglect; that unconscious flashbacks to early fears and destructive parenting could dominate individual and social behavior.

Psychohistory relies heavily on historical biography. Notable examples of psychobiographies are those of Lewis Namier, who wrote about the British House of Commons, and Fawn Brodie, who wrote about Thomas Jefferson.

Areas of study

There are three inter-related areas of psychohistorical study.

1. The history of childhood – which looks at such questions as:

• How have children been raised throughout history
• How has the family been constituted
• How and why have practices changed over time
• The changing place and value of children in society over time
• How and why our views of child abuse and neglect have changed

2. Psychobiography – which seeks to understand individual historical people and their motivations in history.

3. Group psychohistory – which seeks to understand the motivations of large groups, including nations, in history and current affairs. In doing so, psychohistory advances the use of group-fantasy analysis of political speeches, political cartoons and media headlines since the loaded terms, metaphors and repetitive words therein offer clues to unconscious thinking and behaviors.

Independence as a discipline

Psychohistorians have argued that psychohistory is a separate field of scholarly inquiry with its own particular methods, objectives and theories, which set it apart from conventional historical analysis and anthropology. Some historians, social scientists and anthropologists have, however, argued that their disciplines already describe psychological motivation and that psychohistory is not, therefore, a separate subject. Others regard it as an undisciplined field of study, due to its emphasis given to speculation on the psychological motivations of people in history. Doubt has also been cast on the viability of the application of post-mortem psychoanalysis by Freud's followers.

Psychohistorians maintain that the difference is one of emphasis and that, in conventional study, narrative and description are central, while psychological motivation is hardly touched upon. Psychohistorians accuse most anthropologists and ethnologists of being apologists for incest, infanticide, cannibalism and child sacrifice. They maintain that what constitutes child abuse is a matter of objective fact, and that some of the practices which mainstream anthropologists apologize for (e.g., sacrificial rituals) may result in psychosis, dissociation and magical thinking.

Psychogenic mode

Lloyd deMause has described a system of psychogenic modes (see below) which describe the range of styles of parenting he has observed historically and across cultures.

Psychohistorians have written much about changes in the human psyche through history; changes that they believe were produced by parents, and especially the mothers' increasing capacity to empathize with their children. Due to these changes in the course of history, different psychoclasses (or psychogenic modes) emerged. A psychoclass is a type of mentality that results from, and is associated with, a particular childrearing style, and in its turn influences the method of childrearing of the next generations. According to psychohistory theory, regardless of the changes in the environment, it is only when changes in childhood occur and new psychoclasses evolve that societies begin to progress.

The major psychogenic modes described by deMause are:

Mode	Childrearing characteristics	Historical manifestations
Infanticidal	Early infanticidal childrearing: Ritual sacrifice. High infanticide rates, incest, body mutilation, child rape and tortures.	Child sacrifice and infanticide among tribal societies, Mesoamerica and the Incas; in Assyrian and Canaanite religions. Phoenicians, Carthaginians and other early states also sacrificed infants to their gods. On the other hand, the relatively more enlightened Greeks and Romans exposed some of their babies ("late" infanticidal childrearing).
	Late infanticidal childrearing: While the young child is not overly rejected by the mother, many newborn babies, especially girls, are exposed to death.
Abandoning	Early Christians considered a child as having a soul at birth, although possessed by evil tendencies. Routine infanticide was replaced by joining in the group fantasy of the sacrifice of Christ, who was sent by his father to be killed for the sins of others. Routine pederasty of boys continued in monasteries and elsewhere, and the rape of girls was commonplace.	Infanticide replaced by abandonment. Those children who survived the experience did not internalize a completely murderous superego. Longer swaddling, fosterage, outside wetnursing, oblation of children to monasteries and nunneries, and apprenticeship.
Ambivalent	The 12th century saw the first child instruction manuals and rudimentary child protection laws, although most mothers still emotionally rejected their children. Children were often treated as erotic objects by adults.	The later Middle Ages ended abandonment of children to monasteries. Early beating, shorter swaddling, mourning for deceased children, a precursor to empathy.
Intrusive	During the 16th century, particularly in England, parents shifted from trying to stop children's growth to trying to control them and make them obedient. Parents were prepared to give them attention as long as they controlled their minds, their insides, their anger and the lives they led.	The intrusive parent began to unswaddle the infant. Early toilet training, repression of child's sexuality. Hell threats turned into the Puritan child so familiar from early modern childrearing literature. On the other hand, the end of swaddling and wet-nursing made possible the explosive modern takeoff in scientific advance.
Socializing	Beginning in the 18th century, mothers began to enjoy child care, and fathers began to participate in younger children's development. The aim remained instilling parental goals rather than encouraging individuation. Manipulation and spanking were used to make children obedient. Hellfire and the harsher physical disciplinary actions using objects to beat the child disappeared. The Socializing Mode remains the most popular model of parenting in North America and Western Europe to the present day.	Use of guilt, "mental discipline", humiliation, time-out, rise of compulsory schooling, delegation of parental unconscious wishes. As parental injections continued to diminish, the rearing of the child became less a process of conquering its will than of training it. The socializing psychoclass built the modern world.
Helping	Beginning in the mid-20th century, some parents adopted the role of helping children reach their own goals in life, rather than "socialize" them into fulfilling parental wishes. Less manipulation, more unconditional love. Children raised in this way are far more empathic towards others in society than earlier generations.	Children's rights movement, natural childbirth, the abandonment of circumcision, attachment parenting, Taking Children Seriously, unconditional parenting, Parent Effectiveness Training, deschooling and free schooling.

Psychohistorians maintain that the five modes of abusive childrearing (excluding the "helping mode") are related to psychiatric disorders from psychoses to neuroses.

The chart below shows the dates at which these modes are believed to have evolved in the most advanced nations, based on contemporary accounts from historical records. A black-and-white version of the chart appears in Foundations of Psychohistory.

The y-axis on the above chart serves as an indicator of the new stage and not a measurement of the stage's size or relation to the x-axis.

The timeline does not apply to hunter-gatherer societies. It does not apply either to the Greek and Roman world, where there was a wide variation in childrearing practices. The arrival of the Ambivalent mode of child-rearing preceded the start of the Renaissance (mid 14th century) by only one or two generations, and the arrival of the Socializing mode coincided with the Age of Enlightenment, which began in the late 18th century.

Earlier forms of childrearing coexist with later modes, even in the most advanced countries. An example of this are reports of selective abortion (and sometimes exposure of baby girls) especially in China, Korea, Taiwan, Singapore, Malaysia, India, Pakistan, New Guinea, and many other developing countries in Asia and North Africa, regions in which millions of women are "missing". The conflict of new and old psychoclasses is also highlighted in psychohistorians' thought. This is reflected in political contrasts – for instance, in the clash between Blue State and Red State voters in the contemporary United States – and in civil wars.

Another key psychohistorical concept is that of group fantasy, which deMause regards as a mediating force between a psychoclass's collective childhood experiences (and the psychic conflicts emerging therefrom), and the psychoclass's behavior in politics, religion and other aspects of social life.

A psychoclass for postmodern times

According to the psychogenic theory, since Neanderthal man most tribes and families practiced infanticide, child mutilation, incest and beating of their children throughout prehistory and history. Presently the Western socializing mode of childrearing is considered much less abusive in the field, though this mode is not yet entirely free of abuse. In the opening paragraph of his seminal essay "The Evolution of Childhood" (first article in The History of Childhood), DeMause states:

The history of childhood is a nightmare from which we have only recently begun to awaken. The further back in history one goes, the lower the level of childcare, and the more likely children are to be killed, abandoned, beaten, terrorized and sexually abused.

There is notwithstanding an optimistic trait in the field. In a world of "helping mode" parents, deMause believes, violence of any other sort will disappear as well, along with magical thinking, mental disorders, wars and other inhumanities of man against man. Although, the criticism has been made that this itself is a form of magical thinking.

Criticisms

There are no departments dedicated to "psychohistory" in any institution of higher learning, although some history departments have run courses in it. Psychohistory remains a controversial field of study, facing criticism in the academic community, with critics referring to it as a pseudoscience. Psychohistory uses a plurality of methodologies, and it is difficult to determine which is appropriate to use in each circumstance. Yet this "plurality" is quite circumscribed.

In 1973, historian Hugh A. Trevor-Roper dismissed the field of psychohistory entirely in response to the publication of Walter Langer's The Mind of Adolf Hitler. He contended that psychohistory's methodology rested "on a defective philosophy" and was "vitiated by a defective method." Instead of using historical evidence to derive historical interpretations, Trevor-Roper contended that "psycho-historians move in the opposite direction. They deduce their facts from their theories; and this means, in effect, that facts are at the mercy of theory, selected and valued according to their consistence with theory, even invented to support theory."

DeMause has received criticism on several levels. His formulations have been criticized for being insufficiently supported by credible research. He has also received criticism for being a strong proponent of the "black legend" view of childhood history (i.e. that the history of childhood was above all a history of progress, with children being far more often badly mistreated in the past). Similarly, his work has been called a history of child abuse, not childhood. The grim perspective of childhood history is known from other sources, e.g. Edward Shorter's The Making of the Modern Family and Lawrence Stone's The Family, Sex and Marriage in England 1500-1800. However, deMause received criticism for his repeated, detailed descriptions on childhood atrocities:

The reader is doubtless already familiar with examples of these psychohistorical "abuses." There is a significant difference, however, between the well-meaning and serious, if perhaps simplistic and reductionistic, attempt to understand the psychological in history and the psychohistorical expose that can at times verge on historical pornography. For examples of the more frivolous and distasteful sort of psychohistory, see Journal of Psychohistory. For more serious and scholarly attempts to understand the psychological dimension of the past, see The Psychohistory Review.

Recent psychohistory has also been criticized for being overly-entangled with DeMause, whose theories are not representative of the entire field.

Organizations

Boston University offers a Psychohistory course at the undergraduate level and has published course details.

The Association for Psychohistory was founded by Lloyd deMause. It has 19 branches around the globe and has for over 30 years published the Journal of Psychohistory. The International Psychohistorical Association was also founded by deMause and others in 1977 as a professional organization for the field of psychohistory. It publishes Psychohistory News and has a psychohistorical mail order lending library. The association hosts an annual convention.

The Psychohistory Forum publishes the quarterly journal Clio's Psyche. It was founded in 1983 by historian and psychoanalyst Paul H. Elovitz. This organization of academics, therapists, and laypeople holds regular scholarly meetings in New York City and at international conventions. It also sponsors an online discussion group.

In Germany, scientists taking an interest in psychohistory have met annually since 1987. In 1992, the Gesellschaft für Psychohistorie und politische Psychologie e.V. (“Society for Psychohistory and Political Psychology”) was founded. This society issues the Jahrbuch für Psychohistorische Forschung (“Annual of Psychohistorical Research”)

Notable psychohistorians

This section does not cite any sources. Please help improve this section by adding citations to reliable sources. Unsourced material may be challenged and removed. (December 2022) (Learn how and when to remove this message)

• Lloyd deMause, founder of The Institute for Psychohistory.
• Peter Gay, Sterling Professor at Yale University, author.
• Robert Jay Lifton, a psychiatrist specializing in psychological motivations for war and terrorism.
• Jerome Lee Shneidman, Editor of the Bulletin of the International Psychohistorical Association, established the Seminar in the History of Legal and Political Thought and Institutions at Columbia University.
• Vamik Volkan, psychiatrist, psychoanalyst, University of Virginia professor emeritus, peacemaker, and Nobel Prize nominee.
• Fawn Brodie, Professor at UCLA, and historian and biographer of Thomas Jefferson, Joseph Smith, and others.